INDEX

Explanations

words related to strength and longevity

oai_token-act-pair · gpt-3.5-turbo

terms related to durability and sturdiness of materials or products

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

peak

-0.82

OPA

-0.74

reon

-0.74

ffic

-0.70

 Meadows

-0.69

itia

-0.68

ei

-0.68

luence

-0.67

arthy

-0.66

ween

-0.66

POSITIVE LOGITS

 aluminum

0.87

 aluminium

0.87

 durability

0.87

 durable

0.86

 plastics

0.81

 lightweight

0.81

theless

0.80

uminum

0.79

 composite

0.77

 resilient

0.76

Activations Density 0.056%

words related to strength and longevity

terms related to durability and sturdiness of materials or products

No Comments

No Known Activations